In briefUndercoding is not conservative, it is inaccurate, and the pattern behind it is systematic, measurable in your own data, and fixable.

Less clinic, more RVUs

For a stretch I was the busiest person in my department and still losing roughly 20 percent of the work RVUs I had earned, because my notes did not capture what my visits were. When I cut my clinical time to 80 percent and spent the rest on coding and informatics, my annual wRVUs went up: not more patients, not overcoding, accurate coding for the first time.

Undercoding: Why Most Physicians Leave Money on the Table and Don't Know It

The size of the gap: 15 to 30 percent
The estimated share of wRVUs lost to undercoding and documentation gaps in a typical outpatient practice, from published audits and payer analyses. It varies by specialty and setting, so treat it as an estimate; the loss itself is systematic, which is why it shows up in your own data.
The three plays that capture the most missed work
What your coders can and cannot catch
In many systems coders review only a sample of notes, often spending around three minutes on each. They can only code what the note says; nobody can award credit for complexity you carried in your head but did not write down.

Why Providers Undercode

Default codes. A comfortable level for the most common visit type, applied broadly: 99213 for the surgeon's post-op follow-up, 99213 or 99214 for the primary care physician's chronic disease session.

Notes that do not match the work. The visit was complex; the note never says so. The coder assigns the lower code, correctly, given what the note says.

Fear of audit. Level 4 gets coded when level 5 is warranted, and level 3 when level 4 is. But a complex specialty or a disproportionately sick panel can legitimately weight a distribution toward level 4 and level 5. The defense is not a lower distribution, it is documentation that matches the complexity.

Unfamiliarity with add-on codes. G2211, prolonged services, ultrasound guidance add-ons, and modifier 25 are real work and real wRVUs, and many providers were never taught they exist.

Where the Gaps Are Most Common

A Simple Self-Audit

Pull your last 10 established patient E/M visits and count the notes that document two or more chronic conditions actively managed plus a prescription adjusted, then count how many of those you coded 99213. Those were 99214s (the column-by-column check is in Choosing Your Level), and the count is your baseline gap.

Then pull your last 10 visits that included a procedure. For each one without a modifier 25 E/M, read the note: was there a clinical evaluation beyond the procedure itself? If so, that is missed revenue.

That snapshot proves the gap exists. A personal coding dashboard answers the same questions every month, so you know whether it is closing.

Building Your Personal Coding Dashboard: Track Your wRVUs and Find the Gaps

For years my dashboard was an Excel file: my codes, volumes, and wRVUs, updated by hand. Most providers compare the total on their productivity report to their threshold and move on; the distribution of codes in that report is the diagnostic.

What Data You Need

Ask your coding department or practice administrator for a report covering a rolling 3-6 month period:

Most EHR and practice management systems can produce it; if not, ask for a raw export of your charges.

What to Look For: The E/M Distribution

Count your established patient 99212s, 99213s, 99214s, and 99215s and calculate the percentage at each level. Compare that to the national benchmark for your specialty: MGMA publishes specialty-specific E/M distributions annually, and your coding department may have those or your payer's own.

If your 99213 percentage is well above benchmark, that is the primary gap. Pull a sample of those notes and run the self-audit on them, with the data choosing the notes; the criteria that separate a 99213 from a 99214 are in Choosing Your Level.

Procedure Code Analysis and the Modifier 25 Check

For procedure heavy specialties, three questions do most of the work.

Tracking Over Time

Track the distribution monthly. After any intervention, whether a template change, a coding education session, or a workflow change, the effect should be visible within 30-60 days.

Free tracking tools built for this
The online wRVU tracker (a tab of the calculator) logs today's codes from your phone between patients and shows daily, monthly, and yearly totals; your data stays on your device. The downloadable Excel tracker adds a full dashboard, monthly breakdown, code-by-code totals, and a projected annual dollar value at your contract rate. Both come pre-loaded with current wRVU values for common E/M, procedure, and add-on codes and let you add your own; they are starter values from the shared 2026 library, so update them from the live Work RVU Calculator before annual planning. The CPT/RVU Coding Dashboard looks up any CPT code, shows current wRVU values and Medicare rates, compares against benchmarks, and prices a coding pattern change before you invest the documentation effort.

The Work Is Already Done

You already see the patients, make the decisions, write the notes, and do the procedures. The gap between the credit you get and the credit you earned is a gap in documentation and coding, not in clinical effort. Close it by documenting so the system can recognize the work, then watching your own numbers.

Sources and Further Reading for This Chapter

Put it to workLook up any code in the wRVU Calculator
Previous